[Pw_forum] input file for transmission
Manoj Srivastava wrote: > > But that is exactly the trouble I am having in this. The scattering > region should not be periodic like leads, as now we have infinite > scattering region! But PWSCF always has PBC, so we should have a large 3rd > lattice vector to make the scattering region practically finite. But in this way you would have a cluster made of 5Al and one hydrogen, with the Al atoms at the edges disconnected from the leads. You want the charge density at the edge atoms to be as similar as possible to that of the leads (in this case an infinite monatomic chain). Even if the scattering region is computed as a periodic system with PBCs, the KS potential inside the cell is not k-dependent and you can use it to solve the scattering problem with different boundary conditions (not periodic). The k-dependent KS eigenstates obtained from pw.x calculations are not used in the pwcond calculation, only the V_KS(G) in the (super)-cell is needed (super because it needs to be larger than the periodicity of the monatomic wire). > I dont > see any super cell here. a_3 is just 1.875*a_0, where a_0 is lattice > constant. The atomic postions are all in a_0 unit, which makes me believe > that it is a practically infinite system with a_3=1.875*a_0. > >>> So, physically we are solving for an >>> infinite device region, but in the physical setting of a transmission >>> problem leads are semi-infinite and device is finite. Shouldn't we use >>> some kind of vacum, i.e. taking 3rd lattice vector large, which >>> effectively would represent the finite device region? >> >>> Also how much part >>> of the leads should be taken as part of device region, >> I don't understand this point. The leads are conceptually different thing >> than the >> scattering region. The lead is a periodic unit of the "bulk" region (in this >> case an >> infinitely long monatomic wire) and it is used to compute the generalized >> Bloch states, >> which in turn are propagated in the scattering region. > > In the above example, in principle we can have one atom H as scattering > region, and Al wire as left and righ leads, but we have taken few Al atoms > with H and treated it as scattering region. Thats what I > meant by how much part of leads should be taken as scattering region. OK, I got it now. I think I have already replied to this somehow, when speaking about the convergence criterion. The H impurity perturbs the charge density of the wire (with respect to the pristine wire configuration, that you take as the lead region) not only on the Al atoms hosting the hydrogen, but also on further atoms. You have to include in the scattering region as many Al atoms as needed, such that the perturbation on the charge density induced by H on the edge atoms is negligible (you can check this from the CBS, as pointed out in my previous reply). In this way the charge density will not change abruptly when crossing the border between the lead and the scattering region and the KS potential will not have discontinuities. GS -- o o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | |via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o o
[Pw_forum] input file for transmission
Dear Gabriele, Thank you very much quick reply. I have some follow up questions on this. On Tue, 14 Apr 2009, Gabriele Sclauzero wrote: > > Manoj Srivastava wrote: > > Dear PWSCF users and developers, > > I wish to do a transmission calculation and confused about the input > > file. I have a question on example 12 of the package, where transmission > > of monoatomic Al wire with a H atom adsorbed on the side is done. The SCF > > run in the device region is done with (some part of input file is given > > below) > > it is usually called the "scattering region" > > > > > > > > ibrav = 6, > > celldm(1) =12.0, > > celldm(3) =1.875, > > > > where the atomic postions of different atoms is > > Al 0. 0. 0. > > Al 0. 0. 0.375 > > Al-0.02779870 0. .75537515 > > H 0.19269012 0. .9375 > > Al-0.02779870 0. 1.11962485 > > Al 0. 0. 1.5 > > > > So, looking at the z coordinate of above system, we notice that device > > region is periodic with period 1.875. > > You are right, the scattering region is a periodic system, since pwscf always > uses PBCs. > The lenght of the scattering region id 12.0*1.875 a.u. and contains all 5 Al > and the H > impurity. But that is exactly the trouble I am having in this. The scattering region should not be periodic like leads, as now we have infinite scattering region! But PWSCF always has PBC, so we should have a large 3rd lattice vector to make the scattering region practically finite. I dont see any super cell here. a_3 is just 1.875*a_0, where a_0 is lattice constant. The atomic postions are all in a_0 unit, which makes me believe that it is a practically infinite system with a_3=1.875*a_0. > > > So, physically we are solving for an > > infinite device region, but in the physical setting of a transmission > > problem leads are semi-infinite and device is finite. Shouldn't we use > > some kind of vacum, i.e. taking 3rd lattice vector large, which > > effectively would represent the finite device region? > > > > Also how much part > > of the leads should be taken as part of device region, > > I don't understand this point. The leads are conceptually different thing > than the > scattering region. The lead is a periodic unit of the "bulk" region (in this > case an > infinitely long monatomic wire) and it is used to compute the generalized > Bloch states, > which in turn are propagated in the scattering region. In the above example, in principle we can have one atom H as scattering region, and Al wire as left and righ leads, but we have taken few Al atoms with H and treated it as scattering region. Thats what I meant by how much part of leads should be taken as scattering region. > > > Is there some kind > > of convergence criterion? Is it like keep increasing part of lead in the > > device reion till further increase does not substantial change device > > behavior, e.g. Bloch's state? > > There is a main convergence criterion (though I don't understand if you are > actually > refering to this). You have to increase the scattering region, adding more Al > atoms in the > wire, such that the complex band structure with real wave-vectors computed > using the > leftmost periodic unit of the wire included in the supercell (the H impurity > being in the > middle of the s.c.) converges to the band structure of an impurity-free wire, > obtained for > instance from a pwscf calculation (or from a pwcond calculation with a 1 atom > cell > containing an Al atom). > > To do this you can use pwcond with > ... > prefixt='prefix of the scattering region' > bdl=ratio between the lenght of the periodic unit and celldm(1) > ikind=0 > band_file='name of file containing the CBS' > ... > > > then compare the real bands (contained in .re) with those from > pwscf (obtained > using the 1 atom cell). > > Also convergence of the transmission with the lenght of the scattering region > can be used, > but it is quite more cheap to check convergence of CBS (which can also help > to understand > if everything is going fine), and when the CBS of your "bulk" region (leads) > is correctly > reproduced the transmission should be converged as well. > > HTH > > GS > > > > > > > Regards, > > Manoj Srivastava > > Ph.D. student > > Department of Physics > > University of Florida, Gainesville, FL > > > > > > ___ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > -- > > > o o > | Gabriele Sclauzero, PhD Student | > | c/o: SISSA & CNR-INFM Democritos, | > |via Beirut 2-4, 34014 Trieste (Italy) | > | email: sclauzer at sissa.it | > | phone: +39 040 3787 511
[Pw_forum] input file for transmission
Manoj Srivastava wrote: > Dear PWSCF users and developers, > I wish to do a transmission calculation and confused about the input > file. I have a question on example 12 of the package, where transmission > of monoatomic Al wire with a H atom adsorbed on the side is done. The SCF > run in the device region is done with (some part of input file is given > below) it is usually called the "scattering region" > > > ibrav = 6, > celldm(1) =12.0, > celldm(3) =1.875, > > where the atomic postions of different atoms is > Al 0. 0. 0. > Al 0. 0. 0.375 > Al-0.02779870 0. .75537515 > H 0.19269012 0. .9375 > Al-0.02779870 0. 1.11962485 > Al 0. 0. 1.5 > > So, looking at the z coordinate of above system, we notice that device > region is periodic with period 1.875. You are right, the scattering region is a periodic system, since pwscf always uses PBCs. The lenght of the scattering region id 12.0*1.875 a.u. and contains all 5 Al and the H impurity. > So, physically we are solving for an > infinite device region, but in the physical setting of a transmission > problem leads are semi-infinite and device is finite. Shouldn't we use > some kind of vacum, i.e. taking 3rd lattice vector large, which > effectively would represent the finite device region? > Also how much part > of the leads should be taken as part of device region, I don't understand this point. The leads are conceptually different thing than the scattering region. The lead is a periodic unit of the "bulk" region (in this case an infinitely long monatomic wire) and it is used to compute the generalized Bloch states, which in turn are propagated in the scattering region. > Is there some kind > of convergence criterion? Is it like keep increasing part of lead in the > device reion till further increase does not substantial change device > behavior, e.g. Bloch's state? There is a main convergence criterion (though I don't understand if you are actually refering to this). You have to increase the scattering region, adding more Al atoms in the wire, such that the complex band structure with real wave-vectors computed using the leftmost periodic unit of the wire included in the supercell (the H impurity being in the middle of the s.c.) converges to the band structure of an impurity-free wire, obtained for instance from a pwscf calculation (or from a pwcond calculation with a 1 atom cell containing an Al atom). To do this you can use pwcond with ... prefixt='prefix of the scattering region' bdl=ratio between the lenght of the periodic unit and celldm(1) ikind=0 band_file='name of file containing the CBS' ... then compare the real bands (contained in .re) with those from pwscf (obtained using the 1 atom cell). Also convergence of the transmission with the lenght of the scattering region can be used, but it is quite more cheap to check convergence of CBS (which can also help to understand if everything is going fine), and when the CBS of your "bulk" region (leads) is correctly reproduced the transmission should be converged as well. HTH GS > > Regards, > Manoj Srivastava > Ph.D. student > Department of Physics > University of Florida, Gainesville, FL > > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- o o | Gabriele Sclauzero, PhD Student | | c/o: SISSA & CNR-INFM Democritos, | |via Beirut 2-4, 34014 Trieste (Italy) | | email: sclauzer at sissa.it | | phone: +39 040 3787 511 | | skype: gurlonotturno | o o
[Pw_forum] input file for transmission
Dear PWSCF users and developers, I wish to do a transmission calculation and confused about the input file. I have a question on example 12 of the package, where transmission of monoatomic Al wire with a H atom adsorbed on the side is done. The SCF run in the device region is done with (some part of input file is given below) ibrav = 6, celldm(1) =12.0, celldm(3) =1.875, where the atomic postions of different atoms is Al 0. 0. 0. Al 0. 0. 0.375 Al-0.02779870 0. .75537515 H 0.19269012 0. .9375 Al-0.02779870 0. 1.11962485 Al 0. 0. 1.5 So, looking at the z coordinate of above system, we notice that device region is periodic with period 1.875. So, physically we are solving for an infinite device region, but in the physical setting of a transmission problem leads are semi-infinite and device is finite. Shouldn't we use some kind of vacum, i.e. taking 3rd lattice vector large, which effectively would represent the finite device region? Also how much part of the leads should be taken as part of device region, Is there some kind of convergence criterion? Is it like keep increasing part of lead in the device reion till further increase does not substantial change device behavior, e.g. Bloch's state? Regards, Manoj Srivastava Ph.D. student Department of Physics University of Florida, Gainesville, FL
